Search results

The purpose of this paper is to accurately capture the risks which are caused by each road user in time.

The authors proposed a novel risk assessment approach based on the multi-sensor fusion algorithm in the real traffic environment. Firstly, they proposed a novel detection-level fusion approach for multi-object perception in dense traffic environment based on evidence theory. This approach integrated four states of track life into a generic fusion framework to improve the performance of multi-object perception. The information of object type, position and velocity was accurately obtained. Then, they conducted several experiments in real dense traffic environment on highways and urban roads, which enabled them to propose a novel road traffic risk modeling approach based on the dynamic analysis of vehicles in a variety of driving scenarios. By analyzing the generation process of traffic risks between vehicles and the road environment, the equivalent forces of vehicle–vehicle and vehicle–road were presented and theoretically calculated. The prediction steering angle and trajectory were considered in the determination of traffic risk influence area.

The results of multi-object perception in the experiments showed that the proposed fusion approach achieved low false and missing tracking, and the road traffic risk was described as a field of equivalent force. The results extend the understanding of the traffic risk, which supported that the traffic risk from the front and back of the vehicle can be perceived in advance.

This approach integrated four states of track life into a generic fusion framework to improve the performance of multi-object perception. The information of object type, position and velocity was used to reduce erroneous data association between tracks and detections. Then, the authors conducted several experiments in real dense traffic environment on highways and urban roads, which enabled them to propose a novel road traffic risk modeling approach based on the dynamic analysis of vehicles in a variety of driving scenarios. By analyzing the generation process of traffic risks between vehicles and the road environment, the equivalent forces of vehicle–vehicle and vehicle–road were presented and theoretically calculated.

This paper aims to present a method of reproducing multi-object structures from materials of diverse physical properties with the use of models fabricated by means of rapid prototyping (RP) techniques.

A process of modelling complex anatomical structures of soft tissues and bones using mandible models as examples was described. The study is based on data acquired through standard computed tomography. Physical models of examined objects were fabricated with RP technology from a 3D-CAD virtual model.

In the analysis of complex medical issues, beside numerical methods, one can simultaneously make use of experimental tests to verify obtained results. In the case of experimental tests, it is necessary to fabricate physical models with appropriate material properties. RP techniques used in the method ensure accurate reproduction of the external shape of the fabricated model, whereas consecutive stages allow us to construct moulds and create internal structures within a finished model by wax cast models.

The application of a physical RP model makes the identification of medical problem more efficient and the reconstruction of pathological alterations for experimental tests clearer. It prevents the simplification of assumptions to experimental analysis. The approach may reduce costs of fabricating models for experimental studies and offers the possibility of using materials of desired properties.

The approach developed by the authors and presented in this paper was submitted for patent protection as “A Method of Reconstructing Medical Models with Internal Structure and the Use of Materials of Diverse properties” – patent application no. P.398644.

The purpose of this paper is to create a model of role‐based access control (RBAC) for virtual enterprise (VE).

An access control model for security and management of VE is presented by integrating generic structure of VE and applying the principles of RBAC. In addition, the application of the model to a supply chain oriented VE illustrates that a general access control scheme can ensure the running of VE.

A theory base of access control for the realization of the VE is found.

The paper provides a very useful new model of access control for VE. This paper is aimed at researchers and engineers.

The problem of image retrieval and image description exists in various fields. In this paper, a model of content-based image retrieval and image content extraction based on the KD-Tree structure was proposed.

A Random Forest structure was built to classify the objects on each image on the basis of the balanced multibranch KD-Tree structure. From that purpose, a KD-Tree structure was generated by the Random Forest to retrieve a set of similar images for an input image. A KD-Tree structure is applied to determine a relationship word at leaves to extract the relationship between objects on an input image. An input image content is described based on class names and relationships between objects.

A model of image retrieval and image content extraction was proposed based on the proposed theoretical basis; simultaneously, the experiment was built on multi-object image datasets including Microsoft COCO and Flickr with an average image retrieval precision of 0.9028 and 0.9163, respectively. The experimental results were compared with those of other works on the same image dataset to demonstrate the effectiveness of the proposed method.

A balanced multibranch KD-Tree structure was built to apply to relationship classification on the basis of the original KD-Tree structure. Then, KD-Tree Random Forest was built to improve the classifier performance and retrieve a set of similar images for an input image. Concurrently, the image content was described in the process of combining class names and relationships between objects.

This paper aims to improve the alignment accuracy of large components in aircraft assembly and an evaluation algorithm, which is based on manufacture accuracy and coordination accuracy, is proposed.

With relative deviations of manufacturing feature points and coordinate feature points, an evaluation function of assembly error is constructed. Then the optimization model of large aircraft digital alignment is established to minimize the synthesis assembly error with tolerance requirements, which consist of three-dimensional (3D) tolerance of manufacturing feature points and relative tolerance between coordination feature points. The non-linear constrained optimization problem is solved by Lagrange multiplier method and quasi-Newton method with its initial value provided by the singular value decomposition method.

The optimized postures of large components are obtained, which makes the tolerance of both manufacturing and coordination requirements be met. Concurrently, the synthesis assembly error is minimized. Compared to the result of the singular value decomposition method, the algorithm is validated in three typical cases with practical data.

The proposed method has been used in several aircraft assembly projects and gained a good effect.

This paper proposes a method to optimize the manufacturing and coordination accuracy with tolerance constraints when the postures of several components are adjusted at the same time. The results of this paper will help to improve the quality of component assemblies.

The purpose of this paper is to improve the performance of E-nose in the detection of wound infection. Feature extraction and selection methods have a strong impact on the performance of pattern classification of electronic nose (E-nose). A new hybrid feature matrix construction method and multi-objective binary quantum-behaved particle swarm optimization (BQPSO) have been proposed for feature extraction and selection of sensor array.

A hybrid feature matrix constructed by maximum value and wavelet coefficients is proposed to realize feature extraction. Multi-objective BQPSO whose fitness function contains classification accuracy and a number of selected sensors is used for feature selection. Quantum-behaved particle swarm optimization (QPSO) is used for synchronization optimization of selected features and parameter of classifier. Radical basis function (RBF) network is used for classification.

E-nose obtains the highest classification accuracy when the maximum value and db 5 wavelet coefficients are extracted as the hybrid features and only six sensors are selected for classification. All results make it clear that the proposed method is an ideal feature extraction and selection method of E-nose in the detection of wound infection.

The innovative concept improves the performance of E-nose in wound monitoring, and is beneficial for realizing the clinical application of E-nose.

Multi-domain convolutional neural network (MDCNN) model has been widely used in object recognition and tracking in the field of computer vision. However, if the objects to be tracked move rapid or the appearances of moving objects vary dramatically, the conventional MDCNN model will suffer from the model drift problem. To solve such problem in tracking rapid objects under limiting environment for MDCNN model, this paper proposed an auto-attentional mechanism-based MDCNN (AA-MDCNN) model for the rapid moving and changing objects tracking under limiting environment.

First, to distinguish the foreground object between background and other similar objects, the auto-attentional mechanism is used to selectively aggregate the weighted summation of all feature maps to make the similar features related to each other. Then, the bidirectional gated recurrent unit (Bi-GRU) architecture is used to integrate all the feature maps to selectively emphasize the importance of the correlated feature maps. Finally, the final feature map is obtained by fusion the above two feature maps for object tracking. In addition, a composite loss function is constructed to solve the similar but different attribute sequences tracking using conventional MDCNN model.

In order to validate the effectiveness and feasibility of the proposed AA-MDCNN model, this paper used ImageNet-Vid dataset to train the object tracking model, and the OTB-50 dataset is used to validate the AA-MDCNN tracking model. Experimental results have shown that the augmentation of auto-attentional mechanism will improve the accuracy rate 2.75% and success rate 2.41%, respectively. In addition, the authors also selected six complex tracking scenarios in OTB-50 dataset; over eleven attributes have been validated that the proposed AA-MDCNN model outperformed than the comparative models over nine attributes. In addition, except for the scenario of multi-objects moving with each other, the proposed AA-MDCNN model solved the majority rapid moving objects tracking scenarios and outperformed than the comparative models on such complex scenarios.

This paper introduced the auto-attentional mechanism into MDCNN model and adopted Bi-GRU architecture to extract key features. By using the proposed AA-MDCNN model, rapid object tracking under complex background, motion blur and occlusion objects has better effect, and such model is expected to be further applied to the rapid object tracking in the real world.

The purpose of this paper is to attempt to establish the pattern of multi‐objective and multi‐dimensional grey fuzzy forecasting with feedback based on the theories of grey system and fuzzy recognition.

First, according to the given weights, the weighting integrated value of samples were computed. Second, the method of fuzzy recognition with single index was employed to calculate the fuzzy classification of the integrated value. According to the cause analysis, the fuzzy classification of the integrated value is used to compute the weights of indexes. In the same way, repeating the above processes, the weighting integrated value and fuzzy classification with given accuracy are retrieved at the same time. Finally, the authors calculate the correlation coefficient between the weighting integrated values and forecasting objects, according to the principle of maximal relativity, optimizing the weighting integrated value of samples, establishing the fuzzy forecasting pattern, and checking the model's precision. A numeric example is also computed in the last part of the paper.

The results are convincing: not only that the pattern of multi‐objective and multi‐dimensional grey fuzzy forecasting with feedback based is valid, but also the model's applied prediction accuracy is higher, where the test samples' mean forecast accuracy of groundwater dynamic levels is 96.50 percent.

The method exposed in the paper can be used to predict groundwater dynamic levels and even for other similar forecast problems.

The paper succeeds in realising both a prediction pattern and application of predicting groundwater dynamic levels by using the newest developed theories of grey system and fuzzy recognition.

For autonomous vehicles, trajectory prediction of surrounding vehicles is beneficial to improving the situational awareness of dynamic and stochastic traffic environments, which is a crucial and indispensable element to realize highly automated driving.

In this paper, the overall framework consists of two parts: first, a novel driver characteristic and intention estimation (DCIE) model is built to indicate the higher-level information of the vehicle using its low-level motion variables; then, according to the estimation results of the DCIE model, a classified Gaussian process model is established for probabilistic vehicle trajectory prediction under different motion patterns.

The whole method is later applied and analyzed in the highway lane-change scenarios with the parameters of models learned from the public naturalistic driving data set. Compared with other traditional methods, the performance of this proposed approach is proved superior, demonstrated by the higher accuracy in the long prediction horizon and a more reasonable description of uncertainty.

This hierarchical approach is proposed to make trajectory prediction accurately both in the short term and long term, which can also deal with the uncertainties caused by the perception system or indeterminate vehicle behaviors.

According to the grey uncertainty and the connotation of different types weights, the purpose of this paper is to establish the pattern of multi-dimensional grey fuzzy decision making with feedback based on weight vector and weight matrix, and applies this pattern to evaluate the regional financial innovation ability.

At first, this paper analyzes the connotation of financial innovation ability and establishes the evaluation system of regional financial innovation ability. Second, the formula of computing the multi-objective weighted comprehensive value based on weight vector and weight matrix is put forward. In view of the object function with supervised factor and stability coefficient, this paper gives the formulas to compute weight vector and weight matrix. Moreover, the algorithm of the multi-dimensional grey fuzzy decision making pattern with feedback based on weight vector and weight matrix is expressed. At last, this paper uses the presented pattern to evaluate the financial innovation ability of thirty-one provinces in China.

The results are convincing: the development of regional financial innovation is not balanced in China, having obvious spatial clustering feature. The comparisons of evaluation results based on different forms of weights show that the calculating convergence speed of the pattern presented in this paper is fast. The pattern enhances the rationality of the demarcation point between categories, and the convergence within categories, making the evaluation more reasonable.

The method exposed in the paper can be used at evaluating the regional financial innovation ability and even for other similar evaluation problem.

The paper succeeds in realising both the pattern of multi-dimensional grey fuzzy decision making with feedback and evaluating the regional financial innovation ability by using the newest developed theories: weighted grey and fuzzy recognition theory based on weight vector and weight matrix.